Designing for practicality: a personalized and adaptive framework for real-time EMG-based hand motor decoding

Objective.Despite remarkable advances in electromyography (EMG)-based hand motor decoding, developing a practical and reliable decoder for robotic prosthetic hands remains unsolved. This study highlights inter-individual, inter-session, and intra-session variabilities of EMG signals as practical challenges and introduces a novel personalized and adaptive motor decoding framework, designed to mitigate their impact and improve hand motor decoding.Approach.A dataset was collected from twelve participants (8 male, 4 female), incorporating EMG signals from three forearm muscles during 20 repetitions of 9 distinct hand motions. This data was used to conduct a number of tests for analyzing variabilities of EMG signals, followed by the evaluation of the proposed framework using various classifier models, including multi-layer perceptron, support vector machine, convolutional neural network, and Kolmogorov-Arnold network, as well as different feature extraction methods, some of which were suggested in previous studies.Main Results.For feature extraction, a window size of 100 ms proved optimal, balancing the trade-off between time and accuracy. Focusing on EMG signal variabilities, this study highlights the impact of intra-session variability on classification accuracy, alongside inter-individual and inter-session variabilities. For all models, accuracy declines from an initial average of92.33±6.17%to80.56±9.57%after only 17 repetitions without adaptation. However, the framework, which is designed based on unsupervised adaptation, enhances this degradation to88.88±8.72%, achieving statistically significant improvements, regardless of the classifier structure and feature extraction method used.Significance.Considering the variabilities of EMG signals, the proposed framework is modular and integrates components such as a motion classifier and a feature extractor, which can be selected based on suggestions from prior studies. These are extended by additional elements, including a finite-state machine to identify hand rest and action states and manage state transitions, and a Softmax module designed to ensure the consistency of performed motions and minimize the likelihood of misclassification.